MapReduce-based Solutions for Scalable SPARQL Querying
José M. Giménez‐García, Javier D. Fernández, Miguel A. Martínez‐Prieto · WU Research · 2014
The use of RDF to expose semantic data on the Web has seen a dramatic increase over the last few years. Nowadays, RDF datasets are so big and interconnected that, in fact, classical mono-node solutions present significant scalability problems when trying to manage big semantic data. MapReduce, a standard framework for distributed processing of great quantities of data, is earning a place among the distributed solutions facing RDF scalability issues. In this article, we survey the most important works addressing RDF management and querying through diverse MapReduce approaches, with a focus on their main strategies, optimizations and results.